Skip to content

Follow everything with Twarql and sparqlPuSH

June 19, 2010

Milan Stankovic published a nice blog post about following things on the Web. He regards “following” in the broad sense of watching, getting updated about the life and evolution of something or someone.

He wonders if “we could follow events, and get updates about who joined them, who plans on coming. If we could follow places on Foursquare and get updates of who comes there. If we could follow problems on Hypios and see when they get solved. There seems to be many objects to follow and like”, such as metro lines (any closed stations? delayed trains?), clubs (full? anybody I know there?), etc.

He points out that an obvious problem that will arise is that those updates will “create a great deal of overload, so we might have to come up with ways to filter out”.

The Linked Open Data community is already working on adding to the Web descriptions of people, places, clubs, metro stations, etc. such that we can unambiguously refer to them. sparqlPuSH is a protocol implementation that allows users to subscribe to specific updates from an RDF Store or Linked Data server. One could say, for example, that he/she is interested in following any changes to the availability status of a metro line. The user would encode this in a SPARQL query and send it for example to a search service such as Sindice. With sparqlPuSH, at the moment that any changes matching the SPARQL query are made to the store, a notification will be send to the user in real time.

But we don’t need to rely only on following changes to structured data. Valuable information is sometimes updated in textual form (e.g. tweets). Twarql is our framework for extracting information from tweets, using RDF for annotation and enabling stream filtering with SPARQL. With Twarql users can create SPARQL queries focusing on any object or set of objects. So one could request updates for any tweets that mention the metro line of interest. Or another person could request real-time notifications of users that enter a given store (as long as they make their geo location public on Twitter, and they tweet from that store).

(this post is a draft. I plan to come back to it and add actual SPARQL query examples.)

Leave a Comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: